50 Relative rates TOO 150 200 250 Fetal deaths (under 20 weeks gestation) Fetal deaths (20 weeks or more gestation) Neonatal deaths (under 1 mo.) Infant deaths (1-11 mo.) Childhood mortality (1-4 years) Intervals Legend: HHH Less than 4 months $%$3Q4 months less than a year t^vs^J 1 year I 1 2 years F?Z3 3 or more years Figure 11. Variations in relative mortality rates (see Figure 6) with the interval between the termination of one pregnancy and the beginning of the next from gestation through early childhood, Hawaii, 1953. Source: Yerushalmy et al. (31). Eastman's study (62), mentioned earlier, was also retrospective and was based on 5,158 consecutive births (excluding births to primiparas and also "well-attested criminal abortions") delivered at the Johns Hopkins Hospital between late 1936 and mid-1943. He analyzed his data according to interval since the preceding birth: very brief, less than 12 months; brief, 13 to 24 months; moderate, 25 to 48 months; long, more than 48 months. Abortions, stillbirths, premature delivery, and neonatal deaths were all appreciably higher in the very brief interval group. In all other interval groups these rates were much lower and similar to each other. Eastman noted that if a previous delivery were followed by a conception that ended in abortion or premature delivery, both deliveries were more likely to take place within a year. His "very brief was, in other words, artifically exaggerated; so he attributed the differences he had observed to this factor.nd early childhood mortality, when environmental factors would be most important, the association is consistent and almost linear: as intervals increase, chances of survival increase.